Altered motoneuron properties and spinal serotonin contribute to motor deficits in a rabbit hypoxia ischemia model of cerebral palsy

Cerebral palsy (CP) is caused by a variety of factors attributed to early brain damage, resulting in permanently impaired motor control, marked by muscle stiffness and spasticity. To find out if altered physiology of spinal motoneurons could contribute to movement deficits, we performed whole cell patch clamp in neonatal rabbit spinal cord slices after developmental injury at 79% gestation. After preterm hypoxia-ischemia (HI) rabbits are born with muscle stiffness, motor deficits, and increased levels of serotonin (5HT) in the spinal cord. There is a range in severity, thus kits are classified as severely affected, mildly affected, or unaffected based on modified Ashworth scores and other behavioral tests. At postnatal day 0-5, we recorded electrophysiological parameters of 40 motoneurons and a subset of 12 were tested for sensitivity to 5HT. Using a multivariate analysis of neuronal parameters, we found significant differences between severe, mild, unaffected and sham control motoneurons. Severe HI motoneurons showed more sustained firing patterns, depolarized resting membrane potential, and increased instantaneous firing frequency. Interestingly changes in persistent inward currents (PICs) and morphology in severe HI motoneurons do not appear to contribute to excitability. However, severe HI motoneurons were more responsive to αmethyl5HT than sham controls. Since there are higher levels of spinal serotonin in vivo, this would further increase excitability of severe motoneurons and promote muscle stiffness. In summary, changes we observed in spinal motoneuron physiology likely contribute to severity of the phenotype, and therapeutic strategies for CP could target excitability of spinal motoneurons.


Introduction
Cerebral palsy is not well understood, despite its prevalence and seriousness. There exist only a few evidence-based treatments for cerebral palsy: the effectiveness of many currently used therapeutic strategies is unclear (49,64). Part of the problem in treating CP may be the diversity of causes including neonatal stroke, placental insufficiency, preterm birth, inflammation, traumatic injury, difficulties during birth and many other contributing factors (42). Another problem could be that modeling the condition in animals is complicated, and a larger animal models are needed to study motor deficits (15).
Serotonin is generally thought of as a neurotransmitter and neuromodulator, but developmental disruption in 5HT is associated with neurological disorders including autism, Rett syndrome, Down's syndrome and, more recently, cerebral palsy (2,5,18,21,46,62,65,68). In both rodents and rabbits, developmental HI injury results in increased spinal 5HT, which could have a direct impact on excitability and on development of spinal neurons (5,21). Descending serotonergic tracts from the raphe are among the first to arrive in the spinal cord (at 9 weeks post conception in humans (54)), and during periods of activity the spinal cord is bathed in 5HT through volume transmission (12,30). In addition to the developing raphe, other sources of 5HT during development include the maternal raphe and the placenta (7), thus ensuring exposure of the developing fetus to 5HT. Levels of 5HT progressively increase until reaching a developmental peak at 1-3 years of age (35,38). Despite evidence 5HT is important for the fetal nervous system, the effects of altering the presence of 5HT during development are not known. Changes in endogenous 5HT neurotransmission also occur in adulthood after injury and disease and result in profound changes in neuronal activity. After spinal cord injury, muscle spasms are caused by a cascade of activity involving constitutive activity of 5HT receptors and increasing motoneuron (MN) excitability (9,47,48). In summary, altered spinal 5HT could have a profound impact on excitability in CP, and its effects on development could leave a permanent mark on MN excitability.
In order to assess changes in intrinsic properties of spinal motoneurons, we used the rabbit HI model of cerebral palsy (19). It's been shown in previous studies that HI injury during late gestation in rabbits can result in a variety of neurologic and muscular damage, including muscle stiffness (19), loss of neurons in cortical layers 3 and 5, white matter injury, thinning of the corticospinal tract (13), increased spinal 5HT (21), cell death in the spinal cord and decreased numbers of spinal MNs (20), increased sarcomere length, decreased muscle mass and hyperreflexia (56).
Changes in MN physiology are likely to contribute to motor deficits, yet this not been directly assessed in any model of cerebral palsy. In this study we assessed both excitability of the spinal MNs and their ability to respond to 5HT. Both of these factors could dramatically alter excitability in the spinal cord and be potential targets for treatment of cerebral palsy. Our hypothesis was that changes in MN properties would correspond to the severity of motor deficits and altered activity of spinal MNs could contribute to muscle stiffness.

Methods
All rabbits were used according to the University of Rhode Island's Animal Care and Use Committee guidelines. Pregnant New Zealand White rabbits (Charles River Laboratories, Inc, Wilmington MA), underwent HI procedures as described in (19). Briefly, at ~80% gestation (day 25 of gestation (E25) dams were anesthetized, and the left femoral artery was isolated. A Fogarty balloon catheter was inserted into the femoral and advanced to the level of the descending aorta, above the uterine arteries and inflated for 40 minutes. Sham animals underwent the same procedures but without inflation of the catheter. After the procedure, the dam recovered and later gave birth to kits with HI injuries. Categorization of the severity of the phenotype was performed by a blinded observer, using a modified Ashworth scale, observation / tests for activity, locomotion, posture, righting reflex, muscle tone (as described in (19)).

Patch Clamp
Whole cell patch clamp is performed similar to previously published work (50). Briefly, horizontal spinal cord slices 350µm thick are obtained using a Leica 1000 vibratome. Slices are incubated for one hour at 30°C and perfused with oxygenated (95% O 2 and 5% CO 2 ) modified Ringer's solution containing (in mM): 111 NaCl, 3.09 KCl, 25.0 NaHCO3, 1.10 KH2PO4, 1.26 MgSO4, 2.52 CaCl2, and 11.1 glucose at 2 ml/min. Whole cell patch electrodes (1-3 MΩ) contain (in mM) 138 K-gluconate, 10 HEPES, 5 ATP-Mg, 0.3 GTP-Li and Texas Red dextran (150μM, 3000MW). PICs are measured in voltage clamp mode with holding potential of −90mV and depolarizing voltage ramps of both 22.5mV/s and 11.25 V/s bringing the cell to 0mV in 4s, and 8s respectively, and then back to the holding potential in the following 4 or 8 seconds. In current clamp, frequency -current measurements are obtained from current ramps and steps, as well as maximum firing rates and characteristics of action potentials and after-spike after hyperpolarization. Hyperpolarizing current steps are used to test Ih. After electrophysiological measurements are obtained, MNs are imaged to assess anatomical development, and photos are obtained of the electrode placement within the spinal cord slice, as shown in Figure 1. Images are acquired with a Nikon microscope fitted with a 40x water-dipping objective lens and two photon excitation fluorescence microscopy performed with a galvanometer-based Coherent Chameleon Ultra II laser. To optimize excitation of red/green fluorophores, the laser is tuned to 900nm. 3D reconstructions of motoneurons are created using Neurolucida 360° software.
Drug application Serotonergic drugs were bath applied and all electrophysiological parameters were recorded again. Either serotonin was applied at a concentration of 10µM or 0.3 µM α-methyl 5 hydroxy tryptamine in combination with 10 µM citalopram. Since perfusion rate was 2.5 ml/minute and dead space plus the bath volume was 10ml, all recordings were made 20 minutes after application, when the volume of the bath and tubing had been twice replaced.
Statistics A multivariate analysis using SPSS software was used for determining significance of parameters over groups. Injury classification (sham/control, HI unaffected, HI mildly affected, and HI severely affected), age of the kit, and spinal cord region (cervical, thoracic, lumbar or sacral) were all included as fixed factors. Paired T tests were used to determine significance of responses to serotonergic drugs.

Figure 1
Patch clamp of MNs with dye-filling via patch electrode as in shown in (A). Scale bar = 100µm. (B) Placement of patch electrode (at arrow) within the slice is captured with a photo.

HI MNs show sustained firing
In rabbit kits severely injured by HI, MNs had some characteristics of hyperexcitability, including increased sustained firing. The frequency current (F-I) relationship was measured using current ramps, as shown in Figure 2. Depolarizing current ramps are used to evoke firing, and current at onset and offset of firing (I ON and I OFF ) determines ΔI. In sham control MNs, ΔI was larger and always a positive value (191pA +/-160pA), indicating firing ceased at a higher current amplitude than the current level that elicited firing on the ascending ramp (see figure 2A).
Severe HI MNs had a smaller, and sometimes negative ΔI (84pA +/-63pA), revealing increasingly sustained firing (see figure 2B). Figure 2 shows typical MNs from sham control and severe HI kits (black and red traces, respectively). Mean values for these parameters are shown in Figure 2C and D, with the intermediate classifications (HI unaffected and HI mildly affected) combined for brevity of the graphics only. To determine significance of the variables, a multivariate analysis was used that included 3 fixed factors: 1) condition (sham control, HI unaffected, HI mildly affected, and HI severely affected, 2) age (postnatal day 0-5), and 3) spinal region (cervical, thoracic, lumbar or sacral). Significance based on the multivariate analysis is indicated with asterisks (* p < 0.05). All data is included in table format in the supplementary data (Tables S1 -S15). Other significant differences between MNs are shown in Figure 3, including resting membrane potential (RMP) and instantaneous firing rates. Both parameters suggest increased excitability in HI-injured MNs: a depolarized RMP and faster rates of firing. Average resting membrane potential in sham control MNs was -60.3 +/-6.6mV, while in severe HI MNs it was -52.8 +/-5.6mV (p = 0.01). Average instantaneous firing rate in sham control MNs was 98 Hz +/-29 while severe HI MNs could fire at 129 Hz +/-23 (p = 0.02). Sustained firing is also apparent in Figure 3, in the second current step which evokes a brief burst of action potentials followed by depolarization block in both MNs. The severe HI MN recovers and resumes firing while the sham control MN remains in depolarization block. Mean values for these parameters are shown in bar graphs ( Figure  3D, E), with the intermediate classifications (HI unaffected and HI mildly affected) combined. Significance based on the multivariate analysis is indicated with asterisks (* p < 0.05). All data on the active and passive properties is included in the supplementary tables (Table S1 -Table S8).

Persistent inward currents suggest excitability is dampened after HI
PICs were significantly affected by hypoxia-ischemia, but instead of showing signs of hyperexcitability similar to the firing patterns, they revealed that intrinsic excitability may be dampened. Specifically, the voltage dependence was shifted to higher potentials so the maximum amplitude of the PIC in HI severe MNs occurred at a significantly more depolarized voltage compared to sham control MNs, as shown in Figure 4. This would suggest PICs would not be activated as easily in HI injured MNs as their control counterparts. Altered amplitude was not observed outright: the magnitude of the currents was similar (see figure 4D), however the normalized amplitude was significantly smaller in HI MNs (see figure 4E). Normalized amplitude is based on the whole cell capacitance of the MNs, which can be used as an indirect measurement of membrane area. In other words, the current generated per unit of membrane area in HI injured MNs was smaller than in control MNs, and it reached a peak at a higher voltage. These parameters are summarized in Figure 4B -E, with the intermediate classifications (HI unaffected and HI mildly affected) combined only for graphical presentation. Significance based on the multivariate analysis is indicated with asterisks (* p < 0.05; ** p < 0.01). All PIC data is included in supplementary tables S5 and S6, and all data on cell properties is included in supplementary tables S9 and S10. As suggested by the normalized PIC amplitude, membrane properties are affected by HI injury, including a significantly larger capacitance (as shown in Figure 5). Whole cell capacitance is often used as an indirect measure of cell size (larger cells have larger capacitance), thus one might expect increased neuron size after HI injury. However the soma was unchanged in size. No significant changes were observed in soma largest crosssectional area ( Fig  5D), soma volume or soma surface area (data not shown). There was, however a significant increase in dendrite length which could account for the increased capacitance detected in electrophysiological recordings. No changes were measured in the number of dendrites, dendritic nodes  (branching points), dendritic volume or dendritic surface area (data not shown). Significance based on the multivariate analysis is indicated with asterisks (* p < 0.05; ** p < 0.01). All data on neuron morphology is included in supplementary tables S11 -S14.

Motoneuron response to serotonin
Motoneurons were tested with application of serotonin or αmethyl5HT after their electrophysiological parameters had been measured. The response of MNs from kits that had been severely affected by HI was different from the response of MNs from sham animals. For example, PICs from sham control MNs showed a decrease in PIC amplitude while severe HI MNs showed a change in the voltage at onset and the voltage at maximum of the PIC, as shown in Figure 6. Modulation of action potential firing by 5HT also varied between groups. Figure 7 highlights several key parameters that were differently altered, including the current at firing onset (I-ON), the current at firing offset (I-OFF), threshold voltage and the frequency of post synaptic potentials (psps). Significance based on paired T tests (before and after 5HT) is indicated with asterisks (* p < 0.05; ** p < 0.01). All data on changes in active and passive properties before and after 5HT is included in Supplementary Table S15.

Summary
Electrophysiological properties of spinal MNs are altered by developmental HI injury, and the magnitude of changes are correlated to severity of motor deficits. This work also found altered responsiveness of MNs to serotonergic agonist αmethyl5HT. Interestingly, severely affected HI MNs showed a robust increase in excitability in the presence of αmethyl5HT, while sham control MNs had only a weak response to αmethyl5HT. Since traditional views of CP largely view motor dysfunction as a result of the damaged motor cortex improperly signaling to spinal neurons, our new evidence suggests this is only part of the problem. Spinal MNs are not developing the same after HI and show an overall change in excitability, including elevated responses to 5HT compared to sham control MNs. Whether these changes are directly due to the HI insult or indirectly due to downstream effects must be determined by future work.

Contribution of serotonin to motoneuron excitability
Serotonin is a potent neuromodulator of MN activity. Starting early in development, 5HT increases MN excitability, enhances PICs, regulates muscle tone and reflexes, and amplifies synaptic inputs (17,29,31,33,39,59,60,69). Depolarization of the resting membrane potential, increased action potential firing through hyperpolarization of the voltage threshold and enhanced PIC, increased action potential height and reduction of high-voltage activated Ca 2+ entry are all associated with 5HT receptor activation in MNs (3, 22, 25, 32-34, 37, 39, 41). There are seven families of 5HT receptors (5HT 1 -5HT 7 ) containing 14 mammalian receptors. Motoneurons express receptors from each family with the apparent exception only of 5HT 6 receptors (14,16,45,55,67,69). The most well-defined roles in MN excitability are for 5HT 1, and 5HT 2 receptors. Previous studies have shown that activation of 5HT 2A/2B/2C receptors upregulate both Na + and Ca 2+ PICs and increase intrinsic excitability (27,28,36). In the present study we activated the MNs with α methyl 5HT. This is an agonist for 5HT 2/1C receptors, though its effects on MNs are commonly attributed to the more well-established effects of 5HT 2 receptors. Generally 5HT 2 receptors excite MNs while 5HT 1 receptors are inhibitory (36). Changing activity patterns in adults can also alter expression serotonin receptors: after exercise, 5HT 1A receptor mRNA decreased in motoneurons (66), while 5HT immunoreactivity increased (4). Thus the serotonergic system is highly plastic, both during development and in adulthood, in health and after injury.

Developmental role of 5HT
The evolutionary importance of serotoninergic signaling during development is suggested from the existence of several mechanisms which prevent its loss. During gestation in humans, fetal exposure to serotonin arises from maternal sources, from the developing fetal brainstem and from the placenta (6,7). Even when the CNS is isolated from an embryonic mouse, removal of the raphe nucleus is not sufficient to eliminate serotonin: in the absence of serotonergic projections, aromatic acid decarboxylase cells in the developing spinal cord can synthesize 5HT (11), a mechanism that could play a role in serotonergic signaling in the adult spinal cord after descending serotonergic fibers are lost in spinal cord injury (63). In stark contrast to most brainstem-spinal projecting neurons (including rubro-reticulo-and vestibulospinal projecting neurons) which are only capable of axon regeneration at early developmental time points, serotonergic raphe-spinal explants from chick show robust neurite outgrowth and maintained expression of 5-hydroxytryptamine even when obtained well after the permissive period and grown on growth factorfree media (8).
The effects of 5HT on motor circuits are numerous, and dynamic throughout development and adulthood. In the spinal cord, neurons begin expressing 5HT receptors early in embryonic development whether or not 5HT is present (24,69). When 5HT transmission is blocked during development, it alters maturation of inhibitory synapses (1,11). After birth, mouse MNs become much more responsive to 5HT (69), including increasing responsiveness of the PIC to neuromodulation by serotonin (and other neuromodulators) with increasing postnatal age (51). Serotonin also changes neuron outgrowth: 5HT 1A and 5HT 2A receptor activation increased neurite outgrowth, dendritic branching, spine formation and cell survival (10,23,44). In the present study, our finding that dendritic length was increased in HI severe motoneurons is likely the effect of exposure to increased spinal serotonin after HI. In addition to having effects on development, serotonin receptors are themselves affected by development. During postnatal maturation hypoglossal MNs increase expression of 5HT 2A and 5HT 2C receptors, while 5HT 1B receptor expression transiently falls around 2 weeks of age and then increases again (58). After HI injury in rabbits, mRNA for 5HT 2 receptors was reduced in the spinal cord and expression of mRNA for SERT transporters increased (21). It is not yet known if expression of other 5HT receptor subtypes is altered either during normal development in rabbits or after HI injury. Future studies are needed to further investigate these changes.

A role for 5HT in etiology of motor dysfunction
Several conditions including stroke, dorsal rhizotomy, spinal cord injury and traumatic brain injury alter serotonergic signaling, and augmenting serotonin therapeutically could improve recovery (17,26,40,47,48). Acute intermittent hypoxia, which has been found to be therapeutic for some of the above conditions, acts through serotonin receptors (57,61), at least in phrenic motoneurons. Thus, there could be specific effects of both hypoxia and injury that promote upregulation of serotonergic signaling.

Contribution of spinal motoneurons to dysfunction in cerebral palsy
In patients with cerebral palsy, there have been case reports of increased spasticity after starting selective serotonin reuptake inhibitors for treatment of depression (52), and reports of successful treatment of baclofen withdrawal symptoms with serotonergic antagonists (43,53). However exploiting serotonergic signaling pathways for treatment of CP has not been under-explored. Since 5HT is known to largely increase excitability of MNs, its contribution to muscle stiffness and spasticity in cerebral palsy is central to our study. Our results confirm that serotonergic signaling robustly increases MN activity. Previous work that explored intrathecal delivery of serotonergic antagonist methysergide showed that muscle stiffness in rabbits affected by HI was alleviated by blockade of 5HT receptors (21). Here we show that MNs are directly responsive to serotonin, in fact even more so after HI injury. This data supports further exploration of serotonergic modulators, and therapies that are effective at restored balance of serotonergic function for alleviation of spasticity.

Conclusion
Changes in MN physiology after developmental injury are consistent with motor deficits in rabbits. This suggests not only brain injuries but also changes in the spinal cord contribute to impaired function in cerebral palsy. Exploring both altered maturation of spinal neurons and loss of descending connectivity should be pursued to improve outcomes for individuals with cerebral palsy.